@inproceedings{ram2017deepfuse,
title={Deepfuse: A deep unsupervised approach for exposure fusion with extreme exposure image pairs},
author={Ram Prabhakar, K and Sai Srikar, V and Venkatesh Babu, R},
booktitle={Proceedings of the IEEE international conference on computer vision},
pages={4714–4722},
year={2017}
}
论文级别:ICCV 2017
影响因子:-
文章目录
📖论文解读
作者提出了一种解决多曝光融合(MEF)的网络DeepFuse。
这是第一篇使用CNN进行MEF的论文
同时发布了新的基准数据集
🔑关键词
多曝光图像融合,CNN,
💭核心思想
分别将RGB源图像转换为YCbCr,然后分别对每个通道(Y,Cb,Cr)提取特征(增加通道数但是不下采样),通过特征融合(相加),通过特征重构层生成该通道的融合图像,最后将融合后的Y,Cb,Cr转换到RGB生成最终融合图像。
🪢网络结构
网络结构很简单明了,如下图所示。图1黑框内部分的详细说明为图2所示。
📉损失函数
这篇论文太早了,损失函数比较奇怪,我没有慢慢看,有兴趣的同学可以去看原文。
N是总像素数,P是所有像素的集合。
🔢数据集
图像融合数据集链接
[图像融合常用数据集整理]
🎢训练设置
🔬实验
📏评价指标
- MEF SSIM score
参考资料
✨✨✨强烈推荐必看博客 [图像融合定量指标分析]
🥅Baseline
- DeepFuse
🔬实验结果
更多实验结果及分析可以查看原文:
📖[论文下载地址]
🚀传送门
📑图像融合相关论文阅读笔记
📑[GANMcC: A Generative Adversarial Network With Multiclassification Constraints for IVIF]
📑[DIDFuse: Deep Image Decomposition for Infrared and Visible Image Fusion]
📑[IFCNN: A general image fusion framework based on convolutional neural network]
📑[(PMGI) Rethinking the image fusion: A fast unified image fusion network based on proportional maintenance of gradient and intensity]
📑[SDNet: A Versatile Squeeze-and-Decomposition Network for Real-Time Image Fusion]
📑[DDcGAN: A Dual-Discriminator Conditional Generative Adversarial Network for Multi-Resolution Image Fusion]
📑[FusionGAN: A generative adversarial network for infrared and visible image fusion]
📑[PIAFusion: A progressive infrared and visible image fusion network based on illumination aw]
📑[Visible and Infrared Image Fusion Using Deep Learning]
📑[CDDFuse: Correlation-Driven Dual-Branch Feature Decomposition for Multi-Modality Image Fusion]
📑[U2Fusion: A Unified Unsupervised Image Fusion Network]
📚图像融合论文baseline总结
📑其他论文
📑[3D目标检测综述:Multi-Modal 3D Object Detection in Autonomous Driving:A Survey]
🎈其他总结
🎈[CVPR2023、ICCV2023论文题目汇总及词频统计]
✨精品文章总结
✨[图像融合论文及代码整理最全大合集]
✨[图像融合常用数据集整理]
如有疑问可联系:420269520@qq.com;
码字不易,【关注,收藏,点赞】一键三连是我持续更新的动力,祝各位早发paper,顺利毕业~